Current Issue : October-December Volume : 2025 Issue Number : 4 Articles : 5 Articles
Glaucoma is a progressive optic nerve disease and a leading cause of irreversible blindness worldwide. Early and accurate detection is critical to prevent vision loss, yet traditional diagnostic methods such as optical coherence tomography and visual field tests face challenges in accessibility, cost, and consistency, especially in under-resourced areas. This study evaluates the clinical applicability and robustness of three machine learning models for automated glaucoma detection: a convolutional neural network, a deep neural network, and an automated ensemble approach. The models were trained and validated on retinal fundus images and tested on an independent dataset to assess their ability to generalize across different patient populations. Data preprocessing included resizing, normalization, and feature extraction to ensure consistency. Among the models, the deep neural network demonstrated the highest generalizability with stable performance across datasets, while the convolutional neural network showed moderate but consistent results. The ensemble model exhibited overfitting, which limited its practical use. These findings highlight the importance of proper evaluation frameworks, including external validation, to ensure the reliability of artificial intelligence tools for clinical use. The study provides insights into the development of scalable, effective diagnostic solutions that align with regulatory guidelines, addressing the critical need for accessible glaucoma detection tools in diverse healthcare settings....
Dark adaptometry is a non-invasive functional test that assesses the retina’s ability to recover sensitivity in low-light conditions following photobleaching. This review explores the physiological mechanisms underlying dark adaptation (DA), including photopigment regeneration and the critical role of the retinal pigment epithelium in the visual cycle. We detail clinical protocols for dark adaptometry using modern instruments such as the AdaptDx, highlighting methodological advances that improve testing efficiency and reproducibility. The clinical utility of dark adaptometry is examined across a range of inherited and acquired retinal disorders, including age-related macular degeneration (AMD), retinitis pigmentosa (RP), Stargardt disease, diabetic retinopathy (DR), cone–rod dystrophy (CRD), vitamin A deficiency, and congenital stationary night blindness (CSNB). Dark adaptometry has emerged as a sensitive biomarker capable of detecting functional deficits before structural changes are evident, making it a valuable tool for early diagnosis and monitoring disease progression. However, limitations such as age-related variability, patient compliance, and lack of standardization remain challenges to broader clinical adoption. Continued refinement of dark adaptometry protocols and instrumentation is essential to maximize its diagnostic potential in ophthalmic practice....
Background/Objectives: A novel, handheld, standalone device using shape discrimination hyperacuity has been developed to remotely monitor age-related macular degeneration (AMD). Methods: We clinically validated the device in an outpatient dry AMD population to evaluate its usability and comfort. A cross-sectional study was conducted with subjects aged 50 years or older with dry AMD at the University of Michigan Kellogg Eye Center outpatient clinic after approval from the UM IRB (HUM00187177). Subjects used the device and then completed a device survey and System Usability Scale (SUS). Results: Thirty-one subjects completed the study, and one subject withdrew post-study completion (mean age 77 years, STD 8 years). The mean testing time was 126 s (STD 56 s), and the median was 116 s. Most patients reported that the use of the device occurred for an acceptable duration (97%), was easy (97%), and was comfortable (90%). The mean SUS score was 77.7 (STD 11.9). Conclusions: A handheld, standalone device can provide a rapid, easy, and comfortable testing solution for patients with dry AMD. The usability of the device supports further clinical trials to demonstrate the device’s ability to reliably detect the progression of AMD....
Background/Objectives: Certain aspects of retinal thickness assessed by optical coherence tomography (OCT) in patients with Parkinson’s disease (PD) require additional clarification. It is supposed that attributing reduced retinal thickness in PD to dopaminergic loss may not be acceptable as it also happens in diseases where dopaminergic loss does not occur. The objective of our study is to compare the ganglion cell/inner plexiform layer (GCIPL), peripapillary retinal nerve fiber layer (pRNFL), and macular thickness of PD and dopa responsive dystonia (DRD) patients with healthy controls (HC), to investigate whether DRD patients, as a distinctive model of genetically induced dopamine deficiency, have reduced retinal thickness in comparison with PD, and to analyze correlation between retinal thickness and various PD clinical parameters. Methods: We analyzed 86 patients with PD, 10 patients with DRD, and 96 age- and sex-matched HC. Results: GCIPL, pRNFL, and central macula thickness (CMT) are statistically significantly thinner in PD patients compared to HC (p < 0.001, all). GCIPL and CMT are also statistically significantly thinner in DRD patients compared to HC (p = 0.012, p = 0.001, respectively). GCIPL thickness correlates positively with the daily dose of levodopa (r = 0.244, p < 0.01). The thickness of GCIPL and pRNFL correlate negatively with current age (r = −0.219; p < 0.01 and r = −0.358; p < 0.05, respectively). All retinal parameters are statistically significantly thinner in females than in males (p < 0.05). Conclusions: Patients with PD and DRD did not differ in GCIPL and pRNFL thickness when compared to one another. These results, supported by positive correlation of levodopa dose and GCIPL thickness in PD patients, emphasize the importance of dopamine in maintaining retinal thickness....
Background/Objectives: Uncontrolled or long-standing diabetes can lead to proliferative diabetic retinopathy (PDR), a condition that significantly impairs vision. A subset of patients does not respond adequately to conventional therapies, such as intravitreal injections of anti-vascular endothelial growth factor (VEGF) or laser treatment. This study aims to identify potential biomarkers for alternative treatment pathways in the vitreous and blood of patients with severe PDR. Methods: Vitreous samples were collected from PDR patients (n = 3) undergoing vitrectomy for vitreous hemorrhage and from control patients (n = 9) undergoing ocular surgery for epiretinal membrane or macular holes. Blood samples were collected from a separate group of PDR patients (n = 13) and non-diabetic control patients without retinopathy (n = 13). Medical histories were obtained. Two-stage real-time quantitative polymerase chain reaction (qPCR) was used to evaluate mRNA expression levels of genes potentially implicated in PDR, including HIF2A, PAI-1, TIE1, TIE2, ANGPT2, and VEGFA. Molecular and statistical analyses were performed to compare PDR and control vitreous and blood samples. Results: The PDR vitrectomy group included two females and one male, aged 71–77 years (mean 74 years). All participants had undergone pan-retinal photocoagulation and two had received anti-VEGF injections before vitrectomy. These participants had elevated HbA1c levels. Targeted vitreous gene analysis revealed varying levels of increased expression of all genes examined as compared to the control group. A trend for increased median expression was demonstrated for all examined genes: HIF2A by 1.44-fold (PDR = 2.50, control = 1.74, p = 0.21), PAI-1 by 1.56-fold (PDR = 3.00, control = 1.93, p = 0.37), TIE1 by 1.36-fold (PDR = 2.33, control = 1.72, p = 0.66), TIE2 by 2.06-fold (PDR = 2.81, control = 1.36, p = 0.51), ANGPT2 by 2.93-fold (PDR = 6.32, control = 2.16, p = 0.1), and VEGFA by 3.53-fold (PDR = 3.51, control = 0.99, p = 0.08). PDR blood sample analysis as compared to controls showed a trend for increased expression of VEGFA by 1.2-fold (PDR = 0.88, control = 0.74, p = 0.57), whereas the other examined genes showed a trend of reduced expression; HIF2A decreased by 0.50-fold (PDR = 0.38, control = 0.75, p = 0.07), PAI by 0.51-fold (PDR = 0.35, control = 0.69, p = 0.09), TIE-1 by 0.79-fold (PDR = 0.79, control = 1.00, p = 0.54), TIE-2 by 0.70-fold (PDR = 0.58, control = 0.82, p = 0.34), and ANGPT2 by 0.45-fold (PDR = 0.51, control = 1.15, p = 0.11). Conclusions: Vitreous sample analysis revealed a trend of increased mRNA expression of ANGPT2 and VEGFA in patients with PDR. Blood sample analysis did not show a significant increase of VEGFA mRNA expression but a decreased trend of HIF2A, PAI-1, and ANGPT2 mRNA expression. These trends warrant validation in a larger cohort to explore alternative pathways for targeted treatment....
Loading....